R13 REVIEW Estrogens and atherosclerosis: insights from animal models and cell systems Jerzy-Roch Nofer1,2 1Center for Laboratory Medicine, University Hospital Mu¨nster, Albert Schweizer Campus 1, Geba¨ude A1, 48129 Mu¨nster, Germany 2Department of Medicine, Endocrinology, Metabolism and Geriatrics, University of Modena and Reggio Emilia, Modena, Italy (Correspondence should be addressed to J-R Nofer at Center for Laboratory Medicine, University Hospital Mu¨nster; Email: [email protected]) Abstract Estrogens not only play a pivotal role in sexual development but are also involved in several physiological processes in various tissues including vasculature. While several epidemiological studies documented an inverse relationship between plasma estrogen levels and the incidence of cardiovascular disease and related it to the inhibition of atherosclerosis, an interventional trial showed an increase in cardiovascular events among postmenopausal women on estrogen treatment. The development of atherosclerotic lesions involves complex interplay between various pro- or anti-atherogenic processes that can be effectively studied only in vivo in appropriate animal models. With the advent of genetic engineering, transgenic mouse models of atherosclerosis have supplemented classical dietary cholesterol- induced disease models such as the cholesterol-fed rabbit. In the last two decades, these models were widely applied along with in vitro cell systems to specifically investigate the influence of estrogens on the development of early and advanced atherosclerotic lesions. The present review summarizes the results of these studies and assesses their contribution toward better understanding of molecular mechanisms underlying anti- and/or pro-atherogenic effects of estrogens in humans. Journal of Molecular Endocrinology (2012) 48, R13–R29 Introduction circumstances, these hormones promote the develop- ment of atherosclerosis. First, epidemiological studies Sexual hormones are intimately involved in the examining the efficacy of menopausal hormone pathogenesis of atherosclerosis. The strikingly lower therapy demonstrated that E2 applied together with incidence of myocardial infarction (MI) in premeno- progestins increases coronary risk in postmenopausal pausal women than in men in the same age group, women both with and without clinically established together with the dramatic increase in coronary risk coronary heart disease (Hulley et al. 1998, Rossouw after menopause is strongly suggestive of an important et al. 2002). Secondly, no or even positive correlation role of estrogen levels in the etiology of MI in women between plasma estrogen levels and atherosclerosis has (Lerner & Kannel 1986, Pe´rez-Lo´pez et al.2009). been found in male subjects with intact E2 synthesis Studies on males with defective estrogen action add (The Coronary Drug Project Research Group 1970, further support to the notion that these hormones Muller et al. 2004). In addition, administration of E2 in exert potent atheroprotective effects. Actually, acceler- such males has been reported to increase the risk of ated development of atherosclerosis has been observed MI and stroke. While reasons for these profound in male subjects deficient in P450 aromatase, an enzyme discrepancies remain obscure, they may be related to that converts androgens to estrogens, and characterized the following: by estradiol (E2) levels below the detection limit (Maffei et al. 2004). More importantly, estrogen treatment in † divergent estrogen levels – various treatment regi- aromatase-deficient subjects reversed atherosclerotic mens have been used in both clinical and animal changes in the carotid artery. While the above results studies for estrogen substitution. It cannot, there- provide a strong argument for the atheroprotective fore, be excluded that estrogens exert both anti- and effects of estrogens, they are in a remarkable contra- pro-atherogenic effects depending on actual plasma diction to other data suggesting that, under certain hormone concentrations; Journal of Molecular Endocrinology (2012) 48, R13–R29 DOI: 10.1530/JME-11-0145 0952–5041/12/048–R13 q 2012 Society for Endocrinology Printed in Great Britain Online version via http://www.endocrinology-journals.org Downloaded from Bioscientifica.com at 09/26/2021 02:36:05AM via free access R14 J-R NOFER . Estrogens and atherosclerosis † distinct estrogen receptors – estrogens act via distinct on a C57Bl6 background, are characterized by highly pro- receptors including nuclear receptors estrogen atherogenic dyslipidemia with high LDL and low HDL receptor a (ERa) and b (ERb) as well as the recently levels in plasma and develop human-like lesions. identified plasma membrane-bound and G protein- The effects of estrogens on diet-induced athero- coupled GPR30 (GPER). The exact involvement of sclerosis have been studied in atherosclerosis-prone these receptors in processes related to the patho- B6 mice as well as in Apoe KO and Ldlr KO mice under genesis of atherosclerosis has not been fully hypercholesterolemic conditions (Bourassa et al. 1996, explored; Elhage et al. 1997a,b, 2001, Hodgin et al. 2001, 2002, † distinct cell targets of estrogen action – atherosclero- Mayer et al. 2005, Tsuda et al. 2005, Seli et al. 2007, tic lesion arises as a consequence of complex Villablanca et al. 2009). In female animals, the study interplay between lipoproteins, inflammatory design usually encompassed surgical or pharma- mediators, and various cell types (such as macro- cological castration followed by E2 substitution to assure phages, endothelial cells, smooth muscle cells, E2 concentrations in plasma comparable to those seen in B-cells, B-cells, dendritic cells, platelets, myofibro- premenopausal women before ovulation (ca. 100 and blasts, mast cells, etc.). It is conceivable that estrogens 200 pg/ml). Under such experimental conditions, simultaneously exert both pro- and anti-atherogenic ovariectomy exerted no effect or aggravated, while E2 effects depending on their cellular target. substitution almost invariably ameliorated athero- sclerosis. Additional studies documented that varying Examination of vascular lesions in animals represents dose and timing of estrogen replacement critically affect an integral part of experimental approaches helping the disease development. Elhage et al. (1997a)haveshown to study arteriosclerotic processes. With the advent of that low-dose E2 fails to prevent fatty streak formation in genetic engineering, transgenic mouse models have Apoe knockout (KO) mice, while Freudenberger et al. supplemented classical dietary cholesterol-induced (2010) reported even increased atherosclerosis in the disease models such as cholesterol-fed hamster, rabbit, same animal model treated with low E2 concentrations. pig, or monkey. In the last two decades, these transgenic Cann et al. (2008) found that E2 is ineffective in reducing models were widely used to specifically investigate the atherosclerotic burden in Apoe KO mice, when adminis- influence of estrogens on the development of athero- tered 45 days after ovariectomy. The latter observation sclerotic lesions. The present review summarizes the corroborates the results of interventional studies showing results of these studies and assesses their contribution that women undergoing a period of low estrogen levels toward better understanding of molecular mechanisms followed by reintroduction of estrogen treatment are more underlying anti- and/or pro-atherogenic effects of likely to experience a cardiovascular event than those on estrogens in humans. continuous treatment. In addition to E2, natural E2 metabolites or phytoes- trogens also exert anti-atherogenic effects in animal Effects of estrogens on atherosclerosis in models. For instance, 2-methoxyestradiol (2-ME), which animal models is the major metabolite of E2 formed via sequential conversion of E2 to 2-hydroxyestradiol and 2-ME by Atherosclerosis is often defined as a multifactorial cytochrome P450 and catechol-O-methyltransferase disease. While selected pro- or anti-atherogenic pro- (COMT) and is produced in various tissues in addition cesses can be effectively studied under in vitro con- to ovary (Zacharia et al. 2004), was demonstrated to ditions, the complex interplay between them can be attenuate atherosclerosis in female Apoe KO mice explored only in appropriate animal models. Even (Bourghardt et al. 2007). Similar favorable effects were though there is no one perfect animal model that also seen after administration of isoflavones exerting completely replicates the stages of human athero- weak estrogenic effects (Adams et al. 2002a,b). sclerosis, cholesterol feeding is a common feature Anti-atherogenic effects of estrogens in males were shared by most of them. Increased dietary cholesterol studied less intensely than those of females. Available feeding produces a permanent hyperlipidemic state, results indicate, however, that estrogens reduce athero- which favors penetration of pro-atherogenic lipoproteins sclerotic burden in orchidectomized Apoe KO mice on into the arterial wall and produces chronic inflammation cholesterol-rich diet or infused with angiotensin II accompanied by monocytosis promoting the recruitment (Ang-II) as well as in diabetic Apoe KO mice (Elhage of monocytes into the vessel wall. While a cholesterol-rich et al. 1997a, Tse et al. 1999, Martin-McNulty et al. 2003). diet induces the development of atherosclerotic lesions It ought to be emphasized that plasma E2 concen- in the disease-prone C57Bl6 mouse strain,